AI Risk Atlas Prototype/DemoUnofficial independent experiment. Not an official xAI product. Scores can be wrong.

Back to mitigations
In progress

Adversarial honesty evals with hidden goals

Independent suites that reward defection if the model can conceal it. Publish fail cases, not just averages.

$103M experimental capital · 2 sources

Who should own it
AISI / METR / frontier labs
Alliances / evaluators
How quickly it can land
Weeks
A dedicated squad can land it inside two months.
Expedited implementation
6 weeks
45 calendar days with a crash team
Normal implementation
5 months
150 calendar days as a planned program

Risk this mitigates

19
Deceptive agent behavior in the wild

Given that evaluation and production agents can already deceive humans, forge identities, and target live organisations, there is a possibility of those agents generalising deception from the eval harness into real operations resulting in compromised organisations, poisoned logs, and loss of confidence that evaluations measure true model intent.

Residual composite 19 · still above the threshold

Effect if implemented

Applied to every failure scenario on that risk, then re-ranked. Axes are clamped at 1.

Likelihood
1
Consequence
0
Urgency
1